Singing as a figure of speech, music as punctuation: A study into music as a means to support the processing of sung language

Sung language is more difficult to process than spoken language, but still, singing can support the processing of sung texts. This dissertation provides both evidence and an explanation for this paradoxical statement. It presents the Musical Foregrounding Hypothesis (MFH). This hypothesis, further elaborated in a model for lyric processing, states that singing functions in a similar way as a linguistic ‘foregrounding’ device (i.e., as musical foregrounding (MF)): it makes the language stand out, although at the same time it hampers processing it. The MFH was checked against the existing literature, and three sub-hypotheses derived from it were tested: - H1: Singing can enhance language-processing immediately, i.e., at first exposure. - H2: A supportive accompaniment eases language-processing and thus reduces MF. - H3.Aspects of poetic form (such as repetition, accent structure, and segmentation) affect the interpretation and appreciation of a text differently when it is sung from when it is read or declaimed. As a result, music poses genre and situation-dependent constraints on poetic form, but also gives way to possibilities to deploy the interaction between musical and poetic form as a song-specific poetic device (i.e., a foregrounding device). This hypothesis was elaborated by several subsub-hypotheses and a set of preference rules governing the occurrence of repetitions within a song. These hypotheses were tested in two classroom experiments, an EEG experiment, ten online listening experiments, two surveys, two natural experiments, and a corpus analysis of two sets of song lyrics. The results, which are of special interest to education, health care, music industry and poetics, provide evidence for all of them. H1. Presenting a text sung instead of spoken supports recall for the text after the first exposure, particularly in a classroom setting. Furthermore, interactions between lyrics and music can be used to support specific interpretations of the words. For example, words sung to out-of-key notes evoke a relatively large N400 associated with semantic meaning. In line with that, out-of-key notes support non-literal interpretations of ambiguous words sung to them. What’s more, people rate verbatim repetitions of words as more acceptable and more meaningful within a sung text than within a spoken one; the lyrics of sung cabaret songs were rated as less sad and more humorous when they were sung or accompanied by piano music; and timing stressed syllables on-beat supports a singer’s perceived sincerity. H2. Purity of singing and listening comfort were rated higher in accompanied song-versions than in a-cappella ones, whereas the distraction by voiceless intervals between sung or spoken phrases was rated lower. H3. Aspects of song form such as rhyme scheme, stanza length and section order interact with aspects of melody, such as repetition, in a predictable way, which can be related to processing fluency and MF. Such interactions are visible in the popularity of psalms, the performance history of two Dylan songs, listener’s reactions to different renditions of the same song, and the occurrence of specific song forms in different media. These and other findings shed new light on the way we handle song lyrics